Multispectral cameras have been developed for use in various research and industry areas. There are several different ways to implement a multispectral camera, including: 1, with a spectral grating (same as a push-broom hyperspectral camera) 2, with tunable or replaceable filters in front of the lens 3, using filters at the pixel level (similar to an RGB camera, but with more than 3 types of filters) and 4, multiple sensors combined together (commonly used in remote sensing applications). All these methods are either very expensive, slow to take images, or hard to calibrate. Modern smartphones and other handheld devices have incorporated standard RGB cameras, however, they are typically limited in the color bands they can sense and are not typically suitable for multispectral imaging for plant health analysis. Therefore, improvements are needed in the field.